diff --git a/docs/references/http_api_reference.md b/docs/references/http_api_reference.md
index 5fe114a5e5..94d38b4429 100644
--- a/docs/references/http_api_reference.md
+++ b/docs/references/http_api_reference.md
@@ -4542,6 +4542,7 @@ Use this mode for the native agent API.
 - `"user_id"`: `string` (optional)
 - `"return_trace"`: `boolean` (optional, default `false`)
 - `"release"`: `boolean` (optional, default `false`)
+- `"chat_template_kwargs": object` (optional)
 
 #### Streaming events to handle
 
@@ -4643,6 +4644,8 @@ curl --request POST \
   Variables specified in the **Begin** component.
 - `"user_id"`: (*Body parameter*), `string`
   The optional user-defined ID. Valid *only* when no `session_id` is provided.
+- `"chat_template_kwargs"`: (*Body parameter*), `object`  
+  Optional passthrough parameters for the underlying LLM's chat template. Commonly used to toggle thinking/reasoning modes on supported models (e.g., `{"enable_thinking": false}`).
 
 :::tip NOTE
 For now, this method does *not* support a file type input/variable. As a workaround, use the following to upload a file to an agent:
@@ -4714,6 +4717,7 @@ Use the same endpoint and add `"openai-compatible": true`.
 - `"stream"`: `boolean`
 - `"session_id"`: `string` (optional)
 - `"model"`: `string` (optional, accepted for compatibility)
+- `"chat_template_kwargs": object` (optional)
 
 ##### Request examples
 
@@ -4734,7 +4738,10 @@ curl --request POST \
                 "role": "user",
                 "content": "Hello"
             }
-        ]
+        ],
+        "chat_template_kwargs": {
+            "enable_thinking": true
+        }
      }'
 ```
 
@@ -4774,6 +4781,8 @@ curl --request POST \
   Optional existing session ID.
 - `"model"`: (*Body parameter*), `string`  
   Optional compatibility field. The server still routes by `agent_id`.
+- `"chat_template_kwargs"`: (*Body parameter*), `object`  
+  Optional passthrough parameters for the underlying LLM's chat template. Commonly used to toggle thinking/reasoning modes on supported models (e.g., `{"enable_thinking": false}`).
 
 ##### Response
 
diff --git a/docs/release_notes.md b/docs/release_notes.md
index c3e6be245a..bfd921a03e 100644
--- a/docs/release_notes.md
+++ b/docs/release_notes.md
@@ -30,7 +30,7 @@ Released on May 26, 2026.
 - Fixes `/chat/completions` to allow sending only latest message in API payload and removes requirement to transmit full conversation history. [#15197](https://github.com/infiniflow/ragflow/pull/15197) See also [Converse with chat assistant](./references/http_api_reference.md#converse-with-chat-assistant).
 - Weight assigned to vector similarity was not applied during the retrieval phase. [#15108](https://github.com/infiniflow/ragflow/pull/15108)
 - Fixes parser configs not saving on dataset configuration page.
-- Logs wer not fully displayed on data source details page.
+- Logs were not fully displayed on data source details page.
 - Fixes document status filtering failure.
 - Fixes crash guard for empty LLM choices responses.
 - RAG: RAPTOR construction process halted when using the [Infinity](https://github.com/infiniflow/infinity) document engine. [#14998](https://github.com/infiniflow/ragflow/pull/14998)